whisper_pandas.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "6ed224c3-b693-4d9a-a99c-534b98f8414b",
   "metadata": {},
   "source": [
    "# whisper_pandas tutorial\n",
    "\n",
    "This is a basic introduction how to use `whisper_pandas`. It also explains a bit what's in Whisper files and archives along the way.\n",
    "\n",
    "Content (planned):\n",
    "\n",
    "* TBD: Read file, summarise, plots\n",
    "* TBD: Show details how to work with timestamps and values in Pandas\n",
    "* TBD: Show how to write to Parquet\n",
    "* TBD: Create and update file from scratch and show content"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a60bc0a2-0b7d-413f-8a8d-400e69117da2",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import whisper_pandas"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "282adb81-dc59-4caa-a72a-146a7b7d3c66",
   "metadata": {},
   "source": [
    "## Read Whisper file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1c56498c-73c9-429b-8ec8-342d34c79895",
   "metadata": {},
   "outputs": [],
   "source": [
    "wsp = whisper_pandas.WhisperFile.read(\"data/example.wsp.gz\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f4410ef-29f2-4752-8421-31251f4b635a",
   "metadata": {},
   "source": [
    "## Summary metadata and archives"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0b8c96cb-ae64-42aa-aa4f-d6b298de3643",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "WhisperFile(meta=WhisperFileMeta(path='data/example.wsp.gz', aggregation_method='average', max_retention=315363600, x_files_factor=0.5, archives=[WhisperArchiveMeta(index=0, offset=52, seconds_per_point=10, points=1555200), WhisperArchiveMeta(index=1, offset=18662452, seconds_per_point=60, points=5256000), WhisperArchiveMeta(index=2, offset=81734452, seconds_per_point=3600, points=87601)]))"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wsp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "3deb7360-f782-4ee3-8c9d-d8261f6e0c9c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                  value\n",
      "path                data/example.wsp.gz\n",
      "actual size                    21696528\n",
      "expected size                  82785664\n",
      "aggregation_method              average\n",
      "max_retention                 315363600\n",
      "x_files_factor                      0.5\n",
      "\n",
      "         seconds_per_point   points  retention    offset      size\n",
      "archive                                                           \n",
      "0                       10  1555200   15552000        52  18662400\n",
      "1                       60  5256000  315360000  18662452  63072000\n",
      "2                     3600    87601  315363600  81734452   1051212\n"
     ]
    }
   ],
   "source": [
    "wsp.print_info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "7761f254-8381-41f1-b901-80739ccb138e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>path</th>\n",
       "      <td>data/example.wsp.gz</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>actual size</th>\n",
       "      <td>21696528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>expected size</th>\n",
       "      <td>82785664</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>aggregation_method</th>\n",
       "      <td>average</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max_retention</th>\n",
       "      <td>315363600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>x_files_factor</th>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  value\n",
       "path                data/example.wsp.gz\n",
       "actual size                    21696528\n",
       "expected size                  82785664\n",
       "aggregation_method              average\n",
       "max_retention                 315363600\n",
       "x_files_factor                      0.5"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wsp.meta.describe_meta()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5e5ff0ef-c7c7-4825-9685-bed29d61f51f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>seconds_per_point</th>\n",
       "      <th>points</th>\n",
       "      <th>retention</th>\n",
       "      <th>offset</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>archive</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>1555200</td>\n",
       "      <td>15552000</td>\n",
       "      <td>52</td>\n",
       "      <td>18662400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>60</td>\n",
       "      <td>5256000</td>\n",
       "      <td>315360000</td>\n",
       "      <td>18662452</td>\n",
       "      <td>63072000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3600</td>\n",
       "      <td>87601</td>\n",
       "      <td>315363600</td>\n",
       "      <td>81734452</td>\n",
       "      <td>1051212</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         seconds_per_point   points  retention    offset      size\n",
       "archive                                                           \n",
       "0                       10  1555200   15552000        52  18662400\n",
       "1                       60  5256000  315360000  18662452  63072000\n",
       "2                     3600    87601  315363600  81734452   1051212"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wsp.meta.describe_archives()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "5650914a-6acf-4e80-b9d2-d135c1bef8f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: create archive table summary on values and timestamps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "58cf9398-8e3e-4acd-958c-6ef4487641e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: plot archive retention bar charts and where values are filled"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9979f07-70c9-4ff6-8c28-666d76f01609",
   "metadata": {},
   "source": [
    "## Data access\n",
    "\n",
    "The `WhisperArchive.to_frame` method returns the data as a `pandas.DataFrame`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "004433c5-d52e-4e7a-8f8c-51783a01db26",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>timestamp</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>43687</th>\n",
       "      <td>2020-07-29 08:28:10+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43688</th>\n",
       "      <td>2020-07-29 08:28:20+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43689</th>\n",
       "      <td>2020-07-29 08:28:30+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43690</th>\n",
       "      <td>2020-07-29 08:28:40+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70228</th>\n",
       "      <td>2020-08-01 10:11:40+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10991</th>\n",
       "      <td>2021-07-20 13:38:50+00:00</td>\n",
       "      <td>4.077281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10992</th>\n",
       "      <td>2021-07-20 13:39:00+00:00</td>\n",
       "      <td>4.100306</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10993</th>\n",
       "      <td>2021-07-20 13:39:10+00:00</td>\n",
       "      <td>4.123771</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10994</th>\n",
       "      <td>2021-07-20 13:39:20+00:00</td>\n",
       "      <td>4.093601</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10995</th>\n",
       "      <td>2021-07-20 13:39:30+00:00</td>\n",
       "      <td>4.081736</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1555200 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      timestamp     value\n",
       "43687 2020-07-29 08:28:10+00:00  0.000000\n",
       "43688 2020-07-29 08:28:20+00:00  0.000000\n",
       "43689 2020-07-29 08:28:30+00:00  0.000000\n",
       "43690 2020-07-29 08:28:40+00:00  0.000000\n",
       "70228 2020-08-01 10:11:40+00:00  0.000000\n",
       "...                         ...       ...\n",
       "10991 2021-07-20 13:38:50+00:00  4.077281\n",
       "10992 2021-07-20 13:39:00+00:00  4.100306\n",
       "10993 2021-07-20 13:39:10+00:00  4.123771\n",
       "10994 2021-07-20 13:39:20+00:00  4.093601\n",
       "10995 2021-07-20 13:39:30+00:00  4.081736\n",
       "\n",
       "[1555200 rows x 2 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wsp.archives[0].to_frame()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "70af4f10-d7e7-4f47-9b6b-7719da4c9f16",
   "metadata": {},
   "source": [
    "There are several options to control the format, point selection and order:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3ee97940-c0cd-4e77-851f-6e5d8fd1b5ff",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on function to_frame in module whisper_pandas:\n",
      "\n",
      "to_frame(self, dtype: 'str' = 'float64', to_datetime: 'bool' = True, drop_time_zero: 'bool' = True, time_sort: 'bool' = True) -> 'pd.DataFrame'\n",
      "    Convert archive data to pandas.DataFrame.\n",
      "    \n",
      "    Parameters\n",
      "    ----------\n",
      "    dtype : str\n",
      "        Data type for point values\n",
      "    to_datetime : bool\n",
      "        Convert from Unix int timestamps to pandas timestamps?\n",
      "    drop_time_zero : bool\n",
      "        Drop points where time is 0, i.e. that were never filled?\n",
      "    time_sort : bool\n",
      "        Sort points in chronological order?\n",
      "    \n",
      "    Returns\n",
      "    -------\n",
      "    df : pandas.DataFrame\n",
      "        Dataframe with columns \"timestamp\" and \"value\" and the\n",
      "        point position index in the Whisper archive as index.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(whisper_pandas.WhisperArchive.to_frame)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b5547ec-fe6d-410c-9172-63c68b4acf0a",
   "metadata": {},
   "source": [
    "For example here's how to get the full data array, including points that haven't been filled in the Whisper file, in the original order.\n",
    "\n",
    "It shows that the DataFrame index is set to the point index in the Whisper archive, and that in WhisperDB timestamps are Unix second integers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e6d893f9-b815-441c-b72d-0191dea93166",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>timestamp</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1626678420</td>\n",
       "      <td>4.223186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1626678430</td>\n",
       "      <td>4.180295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1626678440</td>\n",
       "      <td>4.188692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1626678450</td>\n",
       "      <td>4.225528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1626678460</td>\n",
       "      <td>4.171471</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1555195</th>\n",
       "      <td>1626678370</td>\n",
       "      <td>4.182014</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1555196</th>\n",
       "      <td>1626678380</td>\n",
       "      <td>4.226593</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1555197</th>\n",
       "      <td>1626678390</td>\n",
       "      <td>4.201265</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1555198</th>\n",
       "      <td>1626678400</td>\n",
       "      <td>4.174173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1555199</th>\n",
       "      <td>1626678410</td>\n",
       "      <td>4.203705</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1555200 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          timestamp     value\n",
       "0        1626678420  4.223186\n",
       "1        1626678430  4.180295\n",
       "2        1626678440  4.188692\n",
       "3        1626678450  4.225528\n",
       "4        1626678460  4.171471\n",
       "...             ...       ...\n",
       "1555195  1626678370  4.182014\n",
       "1555196  1626678380  4.226593\n",
       "1555197  1626678390  4.201265\n",
       "1555198  1626678400  4.174173\n",
       "1555199  1626678410  4.203705\n",
       "\n",
       "[1555200 rows x 2 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wsp.archives[0].to_frame(dtype=\"float64\", to_datetime=False, drop_time_zero=False, time_sort=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb09b0c6-fc74-4255-a35c-9359d3045c7b",
   "metadata": {},
   "source": [
    "If you want a `pandas.Series` of the values with timestamps as index, you can get it like this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "0e072a2c-3392-47cc-89cf-a7e8fe19a0bb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "timestamp\n",
       "2020-07-29 08:28:10+00:00    0.000000\n",
       "2020-07-29 08:28:20+00:00    0.000000\n",
       "2020-07-29 08:28:30+00:00    0.000000\n",
       "2020-07-29 08:28:40+00:00    0.000000\n",
       "2020-08-01 10:11:40+00:00    0.000000\n",
       "                               ...   \n",
       "2021-07-20 13:38:50+00:00    4.077281\n",
       "2021-07-20 13:39:00+00:00    4.100306\n",
       "2021-07-20 13:39:10+00:00    4.123771\n",
       "2021-07-20 13:39:20+00:00    4.093601\n",
       "2021-07-20 13:39:30+00:00    4.081736\n",
       "Name: value, Length: 1555200, dtype: float64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wsp.archives[0].to_frame().set_index(\"timestamp\")[\"value\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c8db5a6c-3ccb-4afb-b1c6-63f4460bd9ff",
   "metadata": {},
   "source": [
    "Here's the second archive:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "c6c8b8b4-0316-40b4-b7ac-256a8831e70b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "WhisperArchiveMeta(index=1, offset=18662452, seconds_per_point=60, points=5256000)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wsp.archives[1].meta"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "bfd5136e-2d9f-48e3-a558-abe11d86e055",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>timestamp</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-02-10 07:07:00+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-02-10 07:08:00+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-02-10 07:09:00+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-02-10 07:10:00+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-02-10 07:11:00+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2331010</th>\n",
       "      <td>2021-07-20 13:35:00+00:00</td>\n",
       "      <td>4.099915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2331011</th>\n",
       "      <td>2021-07-20 13:36:00+00:00</td>\n",
       "      <td>4.104024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2331012</th>\n",
       "      <td>2021-07-20 13:37:00+00:00</td>\n",
       "      <td>4.099772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2331013</th>\n",
       "      <td>2021-07-20 13:38:00+00:00</td>\n",
       "      <td>4.101358</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2331014</th>\n",
       "      <td>2021-07-20 13:39:00+00:00</td>\n",
       "      <td>4.099853</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2331015 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                        timestamp     value\n",
       "0       2017-02-10 07:07:00+00:00  0.000000\n",
       "1       2017-02-10 07:08:00+00:00  0.000000\n",
       "2       2017-02-10 07:09:00+00:00  0.000000\n",
       "3       2017-02-10 07:10:00+00:00  0.000000\n",
       "4       2017-02-10 07:11:00+00:00  0.000000\n",
       "...                           ...       ...\n",
       "2331010 2021-07-20 13:35:00+00:00  4.099915\n",
       "2331011 2021-07-20 13:36:00+00:00  4.104024\n",
       "2331012 2021-07-20 13:37:00+00:00  4.099772\n",
       "2331013 2021-07-20 13:38:00+00:00  4.101358\n",
       "2331014 2021-07-20 13:39:00+00:00  4.099853\n",
       "\n",
       "[2331015 rows x 2 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wsp.archives[1].to_frame()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "be5177d5-e25f-403e-a3de-bf7f253326ab",
   "metadata": {},
   "source": [
    "And the last archive, which in this case is minute-resolution:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "39223b5d-3ed3-4fee-815a-fdb6576a335e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "WhisperArchiveMeta(index=2, offset=81734452, seconds_per_point=3600, points=87601)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wsp.archives[2].meta"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "bf9077da-c4d9-4f5c-821e-234942dcd70b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>timestamp</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-02-10 07:00:00+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-02-10 08:00:00+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-02-10 09:00:00+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-02-10 10:00:00+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-02-10 11:00:00+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38850</th>\n",
       "      <td>2021-07-20 09:00:00+00:00</td>\n",
       "      <td>4.100146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38851</th>\n",
       "      <td>2021-07-20 10:00:00+00:00</td>\n",
       "      <td>4.099958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38852</th>\n",
       "      <td>2021-07-20 11:00:00+00:00</td>\n",
       "      <td>4.099795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38853</th>\n",
       "      <td>2021-07-20 12:00:00+00:00</td>\n",
       "      <td>4.100198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38854</th>\n",
       "      <td>2021-07-20 13:00:00+00:00</td>\n",
       "      <td>4.099755</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>38855 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      timestamp     value\n",
       "0     2017-02-10 07:00:00+00:00  0.000000\n",
       "1     2017-02-10 08:00:00+00:00  0.000000\n",
       "2     2017-02-10 09:00:00+00:00  0.000000\n",
       "3     2017-02-10 10:00:00+00:00  0.000000\n",
       "4     2017-02-10 11:00:00+00:00  0.000000\n",
       "...                         ...       ...\n",
       "38850 2021-07-20 09:00:00+00:00  4.100146\n",
       "38851 2021-07-20 10:00:00+00:00  4.099958\n",
       "38852 2021-07-20 11:00:00+00:00  4.099795\n",
       "38853 2021-07-20 12:00:00+00:00  4.100198\n",
       "38854 2021-07-20 13:00:00+00:00  4.099755\n",
       "\n",
       "[38855 rows x 2 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wsp.archives[2].to_frame()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bae7c21c-8284-4d58-9978-938aba2b254d",
   "metadata": {},
   "source": [
    "## Data analysis\n",
    "\n",
    "Now that we have the data nicely formatted we can use pandas and PyData tools such as e.g. matplotlib or scikit-learn or whatever we like to analyse the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b87b97f6-4ab8-4e78-a261-89bad536eb54",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1080x360 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 5))\n",
    "wsp.archives[2].to_frame().set_index(\"timestamp\").plot(style=\".\", label=\"Archive 2\")\n",
    "wsp.archives[1].to_frame().set_index(\"timestamp\").plot(style=\".\", label=\"Archive 1\")\n",
    "wsp.archives[0].to_frame().set_index(\"timestamp\").plot(style=\".\", label=\"Archive 0\")\n",
    "plt.legend();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "399c0027-7f0d-4583-8d5b-12eec919b600",
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: show a plot already!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e64a203-baa8-4307-8599-69644648a4fa",
   "metadata": {},
   "source": [
    "## Data types and pandas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "88abe60b-e2ed-4838-851d-c08bc86d8a7e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: show details on data types and how to work with timestamps & values in pandas"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8998ca12-3a5f-4234-b12e-4241a7031e23",
   "metadata": {},
   "source": [
    "## Write to Parquet\n",
    "\n",
    "Nowadays Parquet is a very common file format for tabular data. You can use `whisper_pandas` to convert Whisper files to Parquet like this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "c50644c2-6d87-4713-9234-6c30a5bc72e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = wsp.archives[0].to_frame()\n",
    "df.to_parquet(\"data/example_archive0.parquet\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "929a81a3-7f4e-44da-a588-2f874ed68290",
   "metadata": {},
   "source": [
    "If you wanted to, you could also combine the data from multiple archives e.g. like this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "93d7f33a-7c00-4e18-a180-9d8e501e3a5f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>archive</th>\n",
       "      <th>index</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>43687</td>\n",
       "      <td>2020-07-29 08:28:10+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>43688</td>\n",
       "      <td>2020-07-29 08:28:20+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>43689</td>\n",
       "      <td>2020-07-29 08:28:30+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>43690</td>\n",
       "      <td>2020-07-29 08:28:40+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>70228</td>\n",
       "      <td>2020-08-01 10:11:40+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38850</th>\n",
       "      <td>2</td>\n",
       "      <td>38850</td>\n",
       "      <td>2021-07-20 09:00:00+00:00</td>\n",
       "      <td>4.100146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38851</th>\n",
       "      <td>2</td>\n",
       "      <td>38851</td>\n",
       "      <td>2021-07-20 10:00:00+00:00</td>\n",
       "      <td>4.099958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38852</th>\n",
       "      <td>2</td>\n",
       "      <td>38852</td>\n",
       "      <td>2021-07-20 11:00:00+00:00</td>\n",
       "      <td>4.099795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38853</th>\n",
       "      <td>2</td>\n",
       "      <td>38853</td>\n",
       "      <td>2021-07-20 12:00:00+00:00</td>\n",
       "      <td>4.100198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38854</th>\n",
       "      <td>2</td>\n",
       "      <td>38854</td>\n",
       "      <td>2021-07-20 13:00:00+00:00</td>\n",
       "      <td>4.099755</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3925070 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       archive  index                 timestamp     value\n",
       "0            0  43687 2020-07-29 08:28:10+00:00  0.000000\n",
       "1            0  43688 2020-07-29 08:28:20+00:00  0.000000\n",
       "2            0  43689 2020-07-29 08:28:30+00:00  0.000000\n",
       "3            0  43690 2020-07-29 08:28:40+00:00  0.000000\n",
       "4            0  70228 2020-08-01 10:11:40+00:00  0.000000\n",
       "...        ...    ...                       ...       ...\n",
       "38850        2  38850 2021-07-20 09:00:00+00:00  4.100146\n",
       "38851        2  38851 2021-07-20 10:00:00+00:00  4.099958\n",
       "38852        2  38852 2021-07-20 11:00:00+00:00  4.099795\n",
       "38853        2  38853 2021-07-20 12:00:00+00:00  4.100198\n",
       "38854        2  38854 2021-07-20 13:00:00+00:00  4.099755\n",
       "\n",
       "[3925070 rows x 4 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df0 = wsp.archives[0].to_frame().reset_index()\n",
    "df0.insert(0, \"archive\", 0)\n",
    "df1 = wsp.archives[1].to_frame().reset_index()\n",
    "df1.insert(0, \"archive\", 1)\n",
    "df2 = wsp.archives[2].to_frame().reset_index()\n",
    "df2.insert(0, \"archive\", 2)\n",
    "df = pd.concat([df0, df1, df2])\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "459997d1-28c5-4135-8205-a30548195801",
   "metadata": {},
   "source": [
    "Of course you could drop or add any column you like (e.g. \"archive\" or \"index\") or assemble or change the `pandas.DataFrame` in any way you like.\n",
    "Let's drop the `index` column (which was the point index in the Whisper archive), and also let's not write the row index to the Parquet file as an example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b39cad45-5d71-47e9-bc69-1a8d52ca8e4c",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop(columns=[\"index\"]).to_parquet(\"data/example.parquet\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "57dc89f6-7ba5-45ad-ba49-989e169ac099",
   "metadata": {},
   "source": [
    "Like Whisper, Parquet is lossless, so the data comes back bit by bit identical:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "713c87c5-0e41-4e4d-be95-5979607fbb51",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>archive</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2020-07-29 08:28:10+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>2020-07-29 08:28:20+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>2020-07-29 08:28:30+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>2020-07-29 08:28:40+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>2020-08-01 10:11:40+00:00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3925065</th>\n",
       "      <td>2</td>\n",
       "      <td>2021-07-20 09:00:00+00:00</td>\n",
       "      <td>4.100146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3925066</th>\n",
       "      <td>2</td>\n",
       "      <td>2021-07-20 10:00:00+00:00</td>\n",
       "      <td>4.099958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3925067</th>\n",
       "      <td>2</td>\n",
       "      <td>2021-07-20 11:00:00+00:00</td>\n",
       "      <td>4.099795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3925068</th>\n",
       "      <td>2</td>\n",
       "      <td>2021-07-20 12:00:00+00:00</td>\n",
       "      <td>4.100198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3925069</th>\n",
       "      <td>2</td>\n",
       "      <td>2021-07-20 13:00:00+00:00</td>\n",
       "      <td>4.099755</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3925070 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         archive                 timestamp     value\n",
       "0              0 2020-07-29 08:28:10+00:00  0.000000\n",
       "1              0 2020-07-29 08:28:20+00:00  0.000000\n",
       "2              0 2020-07-29 08:28:30+00:00  0.000000\n",
       "3              0 2020-07-29 08:28:40+00:00  0.000000\n",
       "4              0 2020-08-01 10:11:40+00:00  0.000000\n",
       "...          ...                       ...       ...\n",
       "3925065        2 2021-07-20 09:00:00+00:00  4.100146\n",
       "3925066        2 2021-07-20 10:00:00+00:00  4.099958\n",
       "3925067        2 2021-07-20 11:00:00+00:00  4.099795\n",
       "3925068        2 2021-07-20 12:00:00+00:00  4.100198\n",
       "3925069        2 2021-07-20 13:00:00+00:00  4.099755\n",
       "\n",
       "[3925070 rows x 3 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_parquet(\"data/example.parquet\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ab6feaa-26cf-4319-82e5-b1efb2974540",
   "metadata": {},
   "source": [
    "## Create new file and update values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "3c9b6282-b8bf-4daf-9b72-4536771fa56c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: show example how to create a new file and update values\n",
    "# TBD: use whisper for that or implement it in whisper-pandas?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8d054238-cbca-4028-b977-52abe9c19280",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}